"Of all cancers, breast cancer is one of the most sensitive to treatment, resulting in a dramatic mortality decrease in the U.S. in recent years," says Berry. "Frequently, in recent breast cancer history, when we have run studies of adjuvant therapy for node- positive breast cancer, the findings have shown that an innovation indeed delays recurrence and prolongs survival. For example, we've shown that increasing doses of the chemotherapy regimen FAC within the standard dose range improves overall survival and disease-free survival. We've shown the same for the addition of paclitaxel. We've also proved that dose density, in terms of delivery every two weeks versus every three weeks, improves overall survival.

"All of these studies suggest the more you do, the better. It's clear to me that delivering more chemotherapy must benefit some patients. Yet, there is a limit and we seem to have reached a plateau."'

Adjusting for demographics, clinical characteristics, intensity of therapy, estrogen receptor status, Berry thought the study would show a statistical significance in overall survival.

"At a minimum, I thought we would have a chance to identify subsets of patients that would benefit and, with 6,200 patients randomized in the 15 trials, that we would be able to confirm our findings. The fact that we did not identify such subsets in no way lessens the value of our study. It is the definitive study for high-dose breast cancer in the adjuvant setting," Berry says.

To appreciate the importance of the study itself and its findings one must understand the nature of the therapy and its conflicting history.

High-dose chemotherapy can be arduous for the patient: it includes administering very high doses of chemotherapy followed by bone marrow or stem cell transplantation of the patient's own blood stem cells that are collected prior to receiving chemotherapy. The autologous transplant rebuilds the bone marrow, which was effected by the intense chemotherapy, explains Naoto Ueno, M.D., Ph.D., associate professor in M. D. Anderson's Departments of Stem Cell Transplantation and Cellular Therapy and Breast Medical Oncology. While it has become far more tolerable, in the past, the therapy was often debilitating and was associated with a number of serious side effects, including infection, nausea, vomiting, and extreme weakness - sometimes resulting in death from the treatment alone.

"The 1980s and early 1990s represented a period in breast cancer history where more was better in terms of treatment," says Ueno, an author on the study.

Despite its side effects, a few small studies emerged in the early 1990s suggesting that the treatment was beneficial for women with high-risk breast cancer, those with at least ten positive axillary lymph nodes. Yet, these studies were not randomized, the gold standard for testing a therapy, explains Berry. Rather they compared one database to another.

Regardless, breast cancer patients and advocates began demanding the treatment and that the therapy, which averages $100,000, be covered by insurance - an issue that was the subject of many lawsuits, all predicated on transplants being effective. According to Berry, approximately 20,000 women with breast cancer in the U.S. received high-dose chemotherapy.

Most randomized trials of the therapy were initiated in the 1990s and did not confirm earlier positive findings. Rather, most of these trials showed little or no benefit for women with breast cancer. The therapy's stature became even more confusing when data from a large randomized positive trial, presented on the plenary session of the 1999 American Society of Clinical Oncology, was later found to be falsified.

"It was important to do this study so it could be determined if there is any evidence of a benefit to patients, or if there is any subset of patients that benefited from the therapy," Berry says.

For their analysis, the investigators built an expansive database and looked at all 15 trials conducted in the world. More than 6,200 women were enrolled in the studies and were randomized to receive either high-dose chemotherapy (3,118 patients) or additional doses of the standard chemotherapy (3,092 patients).

The median patient follow-up was seven years and the mean age of the women enrolled in the trials was 45. Of the women, hormone receptor status was positive in 55 percent, negative in 28 percent and not available in 17 percent. The decrease in the rate of breast cancer relapse due to higher dose chemotherapy was 13 percent, which was statistically significant. However, the benefit did not translate into survival. The decrease in mortality rate was six percent.

After adjusting for age, trial, hormone status, tamoxifen use and the number of positive lymph nodes, the high-dose regimen slightly improved relapse-free survival by about eight months. However, there was no survival benefit for women who received high-dose chemotherapy.

In the analysis, it was important to focus on the role of dose intensity, says Berry.

"In terms of just how high of a dose the chemotherapy was, some trials used very high doses, while others used more moderated doses, but still high enough to require stem cell support."

Regarding dose intensity, if a study had a big difference in dose intensity between the high and standard dose that was allowed to play a greater role in the investigators' conclusions. If a study had a smaller higher dose or if the standard dose was quite high, then it would not count as much in terms of the comparison between the high- and the standard-dose groups, explains Berry.

"Still, even adjusting for trial and dose differences, we were not able to tease out more than a marginal benefit for greater dose intensity."

The catch, Dr. Berry says, is that the chemotherapies under study are known to be effective. Therefore there has to be some sort of dose effect.

"What we are learning is that the doses used in standard chemotherapy regimens for advanced breast cancer have reached a plateau and that increasing beyond that dose is not delivering a greater benefit," says Berry. "It is likely that there are patients who respond to a specific chemotherapy type. Once you give enough of that chemotherapy to benefit those patients, there's no advantage from additional treatment. You are aiding those that were destined to benefit with moderate doses. The remainder of the patients have tumors that are not sensitive to the chemotherapy being considered and would not benefit no matter how much you give them."